aio

name: aio description: Audit and optimize for AI crawlers, LLM consumption, and answer-engine visibility (Perplexity, Google AI Overview, ChatGPT browse).

/aio - AI Optimization Audit & Fix

When invoked, spawn the aio agent (.claude/agents/aio.md) to perform AIO work.

Sub-Commands

• /aio:audit <app> - Audit an app for AI optimization compliance
• /aio:fix <target> - Fix AIO issues on a specific page or app-wide

Core Principle: If It's Not in the HTML, AI Can't See It

AI crawlers (GPTBot, ClaudeBot, PerplexityBot) do minimal or no JavaScript execution. Content must be in the initial server response.

The Anti-Pattern

page.tsx (server) → imports ClientPage
ClientPage.tsx ('use client') → fetches data, renders everything
Result: AI sees an empty shell

The Correct Pattern

page.tsx (server) → fetches data, renders content in RSC
ClientControls.tsx ('use client') → receives data as props, adds interactivity
Result: AI sees complete content

AIO Focus Areas

Content Visibility (Critical)

• Server-rendered content on all public pages
• No client-page anti-pattern
• No content hidden behind client-side data fetching
• robots.txt allows AI crawlers

Content Structure (High)

• Direct answers in first paragraph (AI snippets pull from opening sentences)
• Question-format headings matching natural queries
• FAQ sections with FAQPage JSON-LD schema
• How-to content with HowTo JSON-LD schema
• Structured lists and tables (not info buried in prose)

Content Quality (Medium)

• Specific data with context (not vague claims)
• Domain terms explicitly defined
• Unique, substantive content per page
• Clear topical focus (one topic per page)

AI Crawler Access

• robots.txt: GPTBot, ClaudeBot, PerplexityBot, Google-Extended explicitly allowed
• llms.txt at domain root (emerging standard)
• No interstitials blocking content from crawlers

High-Value Content Types for AI Citation